Scott Cohen

39

Papers

191

Total Citations

Papers (39)

SURGE: Surface Regularized Geometry Estimation from a Single Image

IMPRINT: Generative Object Compositing by Learning Identity-Preserving Representation

FineMatch: Aspect-based Fine-grained Image and Text Mismatch Detection and Correction

FairDeDup: Detecting and Mitigating Vision-Language Fairness Disparities in Semantic Dataset Deduplication

Refine-by-Align: Reference-Guided Artifacts Refinement through Semantic Alignment

Polarized Color Screen Matting

CompleteMe: Reference-based Human Image Completion

Deep Interactive Object Selection

Two Illuminant Estimation and User Correction Preference

Forecasting Human Dynamics From Static Images

Depth From Defocus in the Wild

Deep Image Matting

Relationship Proposal Networks

MetaShadow: Object-Centered Shadow Detection, Removal, and Synthesis

DVQA: Understanding Data Visualizations via Question Answering

Discriminability Objective for Training Descriptive Captions

When Color Constancy Goes Wrong: Correcting Improperly White-Balanced Images

Deepstrip: High-Resolution Boundary Refinement

PhraseCut: Language-Based Image Segmentation in the Wild

Learning To Predict Visual Attributes in the Wild

GamutMLP: A Lightweight MLP for Color Loss Recovery

TopNet: Transformer-Based Object Placement Network for Image Compositing

ObjectStitch: Object Compositing With Diffusion Model

Beyond White: Ground Truth Colors for Color Constancy Correction

Joint Object and Part Segmentation Using Deep Learned Potentials

MultiSeg: Semantically Meaningful, Scale-Diverse Segmentations From Minimal User Input

Unconstrained Foreground Object Search

AESOP: Abstract Encoding of Stories, Objects, and Pictures

PhraseClick: Toward Achieving Flexible Interactive Segmentation by Phrase and Click

Image Inpainting with Cascaded Modulation GAN and Object-Aware Training

Improving Closed and Open-Vocabulary Attribute Prediction Using Transformers

GALA: Toward Geometry-and-Lighting-Aware Object Search for Compositing

Skeleton Key: Image Captioning by Skeleton-Attribute Decomposition

The Photographer's Eye: Teaching Multimodal Large Language Models to See, and Critique Like Photographers

Effective Learning-Based Illuminant Estimation Using Simple Features

PatchCut: Data-Driven Object Segmentation via Local Shape Transfer

Towards Unified Depth and Semantic Prediction From a Single Image

Interactive Segmentation on RGBD Images via Cue Selection

Object Contour Detection With a Fully Convolutional Encoder-Decoder Network